Experience:

3 years required

Skills:

Software Development, TypeScript, Docker, Python, English

Job type:

Full-time

• Design and develop LLM-based applications and AI agents / agentic workflows using LangGraph and LangChain.
• Build and maintain AI solutions on the Dify platform, including custom tools, workflows, and knowledge pipelines.
• Design and improve RAG (Retrieval-Augmented Generation) systems, covering chunking, embedding, retrieval strategy, and evaluation.
• Perform prompt engineering and systematically optimize LLM output quality (evaluation, testing, guardrails).
• Integrate AI solutions with internal systems via REST API / Webhook / MCP.
• Deploy, monitor, and maintain AI systems in production, including managing cost, latency, and token usage.
• Collaborate with Product / Business teams to translate business problems into practical, production-ready AI solutions.
• Stay current with emerging AI/LLM technologies, build PoCs, and propose suitable approaches to the team.
• Bachelor's degree or higher in Computer Science, Computer Engineering, or a related field (or equivalent experience).
• At least 3 years of software development experience with TypeScript or Python.
• Hands-on experience building at least one real-world LLM application (e.g., chatbot, RAG, AI agent) used in production.
• Experience with frameworks such as LangChain, LangGraph, or equivalents (e.g., LlamaIndex, Semantic Kernel).
• Solid understanding of key LLM concepts such as context window, embedding, vector search, function/tool calling, and structured output.
• Experience working with REST APIs, Docker, and Git.
• Able to read technical documentation in English proficiently.
• Nice to Have.
• Experience using or maintaining the Dify platform (or other low-code LLM platforms such as Flowise, n8n).
• Experience with vector databases such as Qdrant or pgvector.
• Knowledge of LLM evaluation / observability tools such as LangSmith or Langfuse.
• Understanding of MCP (Model Context Protocol) or tool integration for AI agents.
• Experience fine-tuning or serving open-source models (e.g., via vLLM, Ollama).
• Basic frontend knowledge (React/Next.js) for building demos or internal tools.
• Core Technology Stack.
• Languages: Python, TypeScript.
• AI Frameworks: LangChain, LangGraph (LlamaIndex / Semantic Kernel equivalents accepted).
• AI Platform: Dify / n8n.
• RAG & Vector DB: Qdrant, pgvector, embedding & retrieval pipelines.
• Evaluation & Observability: Middleware, LangSmith, Langfuse (bonus).
• Integration: REST API, Webhook, MCP.
• Deployment: CI/CD, Git, Docker.
• Model Serving: vLLM, Ollama (bonus).
• Frontend: React / Next.js (bonus).